1. Field of the Invention
This invention relates to fluid flow measuring apparatus. More particularly, this invention relates to flow meters of the vortex shedding type, typically employed to measure the velocity of fluid flow (either liquid or gas) through a pipe or other channel.
2. Description of the Prior Art
It has been known for many years that vortices are developed in a fluid flowing past a non-streamlined obstruction. It has also been known that with certain arrangements the vortices are developed by alternately shedding at regular intervals from opposite edges of the obstruction to form corresponding rows of vortices. Such vortices establish a so-called von Karman "vortex street", which is a stable vortex formation consisting of two nearly-parallel rows of evenly-spaced vortices travelling with the flow stream.
In a von Karman vortex street, the vortices of one row are staggered relative to those of the other row by approximately one-half the distance between consecutive vortices in the same row. The spacing between successive vortices in each row is very nearly constant over a range of flow rates, so that the frequency of vortex formation is correspondingly proportional to the velocity of the fluid. Thus, by sensing the frequency of vortex shedding, it is possible to measure the fluid flow rate.
Various proposals have been made for such flow measuring apparatus of the vortex-shedding type, and some equipment has gone into commercial use. Commonly, such apparatus comprises a rod-like vortex-shedding obstruction positioned in the flowing fluid at right angles to the direction of fluid flow. The obstruction has in many suggested arrangements been a right-circular cylinder, typically a relatively thin, elongate element as shown for example in U.S. Pat. No. 3,564,915 (FIG. 4). Other shapes have been proposed. For example, U.S. Pat. No. 3,116,639 (Bird), shows in FIG. 10 an obstruction of triangular cross-section positioned with one flat surface facing upstream. In like vein, U.S. Pat. No. 3,572,117 (Rodely) also shows the same triangular cross-section arrangement, and additionally shows various other shapes comparable to known configurations as disclosed for example in "Fluid Dynamic Drag", published in 1965 by S. F. Hoerner (see particularly pages 3-7 and 3-17).
A number of different techniques have been proposed for detecting the shedding vortices so as to devlop a flow signal responsive to the shedding frequency. Thermal sensors of the so-called "hot-wire" type (i.e., thermistors, hot films, etc.) frequently have been used in vortex flow meters. The electrical resistance of such sensor elements varies with changes in the cooling rate caused by the passage of the vortices, or by changes in streamline velocity, and this resistance variation is detected by measuring the corresponding changes in current flow through the element.
Such thermal detectors have not been satisfactory for industrial applications. The sensor elements typically are delicate and subject to damage from wear or impact, and also are subject to shorting-out from fluid leakage. A potential hazard is created because the sensor elements must be heated to a temperature above that of the flowing fluid, and because an electrical current must be introduced into the sensor equipment. The output signal generally is small and difficult to detect without highly complex electronic circuitry.
In addition, the output signal appears as a change-in-level of a non-zero current, and thus inherently presents a problem of separating the variable component from the fixed signal level. The output signal variation ordinarily is a small fraction of the fixed signal level, and is particularly subject to noise due to cooling effects from sources other than vortices, as well as being subject to extraneous variations resulting from changes in ambient conditions. Moreover, the output signal variations decrease with increasing vortex frequency, and thus tend to be lost in noise signals at the higher flow rates. Protective coatings on the sensor element are generally quite thin in order to minimize this effect, but this, in turn, results in undesirably low resistance to wear from the flowing fluid.
Other types of detectors have been suggested in an effort to overcome the deficiencies of thermal sensors. In one vortex-detecting arrangement used commercially, a shuttle-like element is mounted in a lateral passageway through the vortex-shedding obstruction to be oscillated back-and-forth by the pressure fluctuations of the passing vortices. The shuttle movement is detected by a nearby pick-up coil to produce a signal reflecting the frequency of vortex generation. The above-mentioned U.S. Pat. No. 3,564,915 shows such an oscillating type of sensor using a ball element (FIG. 7B). That patent also suggests (FIG. 7A) a diaphragm-type device mounted in the center of a lateral bore extending through a rod-shaped obstruction, but apparently does not relate this to any particular sensor design.
As still another approach to the problem, the above-mentioned Bird U.S. Pat. No. 3,116,639 shows a relatively thin vane-like sensing element located downstream of the vortex-shedding obstruction, positioned in alignment with the direction of fluid flow and centrally located so that the spaced rows of vortices pass along opposite sides thereof. This vane-like element is said to oscillate rotatably in a twisting, torsional movement about an axis perpendicular to the fluid flow direction, in response to the pressure fluctuations of the vortices passing thereby. It is proposed in the patent that the length of the vane, in the direction of fluid flow, should be equal to the vortex spacing in a row of vortices.
Various electrical transducer means are proposed in the above-mentioned Bird U.S. Pat. No. 3,116,639 for detecting the intended rotational movement of the vane, as by sensing with conventional electro-magnetic means the oscillatory twisting motion of a support shaft for the vane. This patent also puts forth the notion that the vane it discloses might be made of a piezo-electric material which is strained cyclically by the passage of the vortices along its operative faces to produce an alternating voltage. Piezo-electric means also are proposed to be used as a fluid-fluctuation detectors in U.S. Pat. Nos. 2,809,520 and 3,218,852. None of these prior disclosures, however, shows a practical flow meter arrangement, and developers of commercially-offered apparatus have not attempted to adapt piezo-electric devices to vortex flow meters, resorting instead to other arrangements such as the thermal detectors described previously herein.
The vortex flow-metering devices proposed heretofore have suffered from important drawbacks. For example, certain types of such flow meters have been generally complex and quite expensive to manufacture, and thus have not been adaptable to many applications where cost is a significant factor. Also, available vortex flow-metering apparatus typically has not been adequately reliable in operation, and particularly in some cases has not been capable of satisfactory operation for measuring the flow rate of hostile fluids, e.g. fluids containing dirt, corrosive liquids or gases, or other potentially harmful material such as material which tends to coat the surfaces of an object in the flow stream. Many vortex flow meters have not been capable of sufficiently linear operation over desirably wide ranges of fluid flow rates.